Study on Advanced Key Technologies for Land Use and Land Cover Change Detection Using Remotely Sensed Imagery

Principal Investigator: Dr. Zhang Lu

The overall objective of the project is to develop a suite of new methods to address several key problems in land use land cover change detection using remotely sensed image data. These methods include: (1) a new radiometric data normalization algorithm that is robust to outliers; (2) a novel image segmentation method for changed area extraction from difference image; (3) an improved image classification scheme combining SVM and MRF for change type identification; (4) a new strategy for change detection using SAR images by synthetic analysis of multiple image features. These methods can form a complete work flow for remote sensing change detection, and will be applied to detect and quantify LULC changes in the Great Pearl River Delta region over the past decades.